The impact of climate change on small municipal water resource management The case of Bredasdorp, South Africa - PIERRE MUKHEIBIR
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
The impact of climate change on small
municipal water resource management
•
The case of Bredasdorp, South Africa
PIERRE MUKHEIBIR
May 2007
ENERGY RESEARCH CENTRE
University of Cape TownImpact of climate change on municipal water resources ii
Abstract
In response to the recent reports on climate change impacts, especially on the water supply sector, a
methodology that incorporates climate change impacts in to water resource planning at small
municipal level is proposed. The introduction of the term Climate Impact Factor helps to quantify
the extend to which future drying will have on local water resources. The illustrative example
reveals that a 30% reduction of available water by 2035 will be caused by a projected 8% reduction
in precipitation. The reduction in available water resources due to climate change leads to in an
increase of future supply cost of 3.5 times that under historical climate conditions. It is illustrated
however, that this can to some extend be mitigated through demand side management strategies.
KEY WORDS:
climate impact factor, climate change, groundwater recharge, adaptation, water resource
management
Acknowledgements
This study was made possible with funding from the Korean Environment Institute.
Publication
A copy of this paper was sent to the journal: “Water Research” for publication in May 2007. Still
awaiting approval
Words: 8029
Author details:
Pierre Mukheibir is a Researcher at the Energy Research Centre at the University of Cape Town,
working on adaptation to the impacts of climate variability and change, specifically in water
resources management at municipal level.
Address: Energy Research Centre, University of Cape Town, Private Bag, Rondebosch 7700, South
Africa; tel: +27 21 6502824; fax: +27 21 6502830; e-mail: Pierre.Mukheibir@uct.ac.za
ENERGY RESEARCH CENTREContents
1. Introduction and background 1
2. Overview of current South African water sector 2
2.1 Water is a scarce resource 2
2.2 Equitable access to water 3
3. Projected climate change scenarios for South Africa 4
4. Vulnerability to climate variation and change in the water sector 4
5. Methodological approach to water planning for climate change impacts 5
5.1 Forecast of future water use and demand 6
5.2 Projected climate forecasts 6
5.3 Establish the climate impact on water resources 6
5.4 Water resource management strategies 7
6. Water balance assessment for Bredasdorp 9
6.1 Present water resources 9
6.2 Future supply options 10
6.3 Water demand 10
6.4 Unaccounted for water – Losses 10
7. Impact of climate change on water resource management 11
7.1 Projected annual precipitation for the Bredasdorp catchment 11
7.2 Run-off estimation 11
7.3 Recharge estimation 12
7.4 Impact on water supply 13
8. Analysis of potential water resource management strategies under climate change
impacts 14
8.1 Selected supply-side options 14
8.2 Selected demand-side strategies 15
9. Financial assessment of strategy options 16
10. Conclusion 18
References 19List of figures
Figure 1: Water demand for 2000 per sector ........................................................................................3
Figure 2: Diagrammatic view of the linkage between climate variability and climate change ............7
Figure 3: Annual precipitation for Bredasdorp ...................................................................................11
Figure 4: Water supply and demand graph under historical and climate change conditions for
Bredasdorp ..........................................................................................................................................13
Figure 5: Timing of projected water supply shortfall for Bredasdorp ................................................13
Figure 6: Projected water supply till 2035 under normal climate conditions .....................................14
Figure 7: Projected water supply till 2035 under climate change conditions .....................................14
Figure 8: Projected water supply till 2035 under climate change and DSM conditions .....................16
Figure 9: Magnitude of the increase in unit cost due to climate change over time.............................17
List of tables
Table 1: Available yield in year 2000...................................................................................................2
Table 2: Project annual population growth ...........................................................................................6
Table 3: Actual abstraction ...................................................................................................................9
Table 4: Water abstracted from Klein Sanddrift Dam ..........................................................................9
Table 5: Summary of Bredasdorp water supply (in kl/year)...............................................................10
Table 6: Future groundwater supply options ......................................................................................10
Table 7: In-line water losses ...............................................................................................................10
Table 8: Future groundwater supply options commissioned (in kl/year)............................................15
Table 9: Examples of demand-side options ........................................................................................15
Table 10: Capital cost comparison of ensuring water supplies under climate change conditions ......16
Table 11: Capital cost comparison of ensuring water supplies under climate change conditions
and DSM .............................................................................................................................................17Impact of climate change on municipal water resources 1 1. Introduction and background When faced with actual or impending water shortages, water authorities throughout the world tend to estimate future requirements by extrapolating past trends in consumption, adjusted for expected increase in population and industrial growth. Water planning and management relies on the assumption that the future climate will be the same as the historical trends and hence all our water- supply systems are designed with this assumption in mind. Dams are sized using available information on existing flows in rivers and historical rainfall figures. The most detailed and sophisticated planning methods in use in the urban water sector treat weather as an uncertain-but-stationary process. In other words, climate is assumed to be fixed. But climate does change – it has in the past and it will in the future. Many uncertainties remain about the timing, direction and extent of the climatic changes, as well as the implications. The most important effect of climate change for water supply systems is the increase in uncertainty, which greatly complicates rational water resource planning (Gleick 1998). Long-term demand forecasts have generally used simplistic methods which ignore all influences on water use except population, and in most cases have overstated future water use. Water supply, while recognised as uncertain in the short term, has been assumed stationary in the long term. Future supply is expected to be the same on average as current supply. Supply works are designed for drought intervals of 50 years or more. There is considerable margin for error in the early years of a new facility. Overcapacity, often invisible to the public, has long been preferred to the possibility of undercapacity with its associated political risks. Least cost planning which is so familiar to electric utilities, is hardly heard of in the water industry. However in the latter half of the 20th century, difficulties in meeting planned supply targets were experienced due to diminishing numbers of untapped resources, competition with other users, or constraints imposed by environmental policy. As a result, sharp price increases lowered water use levels much below forecasts, which has led to a re-examination of forecasting practices (Boland 1998). South Africa is an arid country and therefore water is a limiting resource for development and a change in available water supply could have major implications in most sectors of the economy. Factors that contribute to vulnerability in water systems in southern Africa include seasonal and inter-annual variations in rainfall, which are amplified by high run-off production and evaporation rates. Current modelling scenarios suggest that there will be significant climate change impacts in South Africa, even given a business-as-usual global emissions scenario. Climate change is expected to alter the present hydrological resources in southern Africa and increase the need for adaptation to climate variability of future water resources. It has been projected that by mid-century annual average river run-off and water availability will decrease by 10-30% in some dry regions at the mid- latitudes and dry tropics (IPCC 2007). During the past 20 years, most of Africa has experienced extensive droughts, the recent ones being attributed to El Niño. If the occurrence of drought becomes more frequent, the impact on water resources would be significant. Based on these projections, the most severe impacts are likely to occur along the western part of South Africa. The South African Climate Change Response Strategy (DEAT 2004) states that South Africa’s rainfall is already variable in spatial distribution and unpredictable, both within and between years. Much of the country is arid or semi-arid and subject to droughts and a reduction in the amount or reliability of rainfall would exacerbate the already serious lack of surface and groundwater. Therefore it is necessary that the management of water resources at local level needs to be closely integrated into the development plans of local governments. Climate change models are not predictions of the future, but are rather projections of how the future global and local climates may evolve and how these scenarios could affect such things as local water resources. It is therefore important that planners, investors and decision-makers take into account the potential effects of climate change on the water resources and adopt strategies that ensure the long- term sustainability of the water supplies and the local resources. In this paper we discuss both climate change and climate variability, so it is important to clarify the distinction between the two. Climate variability can be thought of as the way climatic variables (such as temperature and precipitation) depart from some average state, either above or below the average value. Although daily weather data depart from the climatic mean, the climate is considered to be stable if the long-term average does not significantly change. On the other hand, climate change can ENERGY RESEARCH CENTRE
Impact of climate change on municipal water resources 2
be defined as a trend in one or more climatic variables characterized by a fairly smooth continuous
increase or decrease of the average value during the period of record.1
There is little in the way of concrete examples of adaptation and more specifically the costs of
adaptation in the municipal water sector. This paper firstly evaluates the climate impacts for a
specific municipal area, Bredasdorp, with regard to its surface and groundwater resources. The
additional water supply required to meet the projected future demand in 2035 is calculated and the
related cost implications are discussed. This process outlines a methodology that can be followed by
other municipal engineers and planners when planning future water management strategies.
2. Overview of current South African water sector
2.1 Water is a scarce resource
South Africa is a water-stressed country with an average annual rainfall of 500mm (60% of the
world average). Only a narrow region along the south-eastern coastline receives good rainfall, while
the greater part of the interior and western part of the country is arid or semi-arid. 65% of the
country receives less than 500m per year, which is usually regarded as the minimum for dryland
farming; 21% receives less than 200mm per year (DWAF 1994).
The natural availability of water across the country is variable, and rainfall displays strong
seasonality. Stream flow in South African rivers is at a relatively low level for most of the year. This
feature limits the proportion of stream flow that can be relied upon for use. Moreover, as a result of
the excessive extraction of water by extensive forests and sugar cane plantations in the relatively
wetter areas of the country, only 9% of the rainfall reaches the rivers, compared to a world average
of 31% (DWAF 1996). Rainfall variability also has implications for water-related disasters such as
floods and droughts.
Many urban and industrial developments, as well as some dense rural settlements, have been
established in remote locations at a distance from adequate reliable water sources. As a result, the
requirements for water already far exceed the natural availability of water in several river basins, and
therefore large-scale transfers of water across catchments have been implemented, like the Lesotho
Highlands Water Scheme. Groundwater also has an important role to play in rural water supplies,
but few major groundwater aquifers exist that can be utilised on a large scale due to high salinity in
most parts of the country (DWAF 1994).
The total average annual available water in South Africa is 49 200x106m3 (this includes the inflow
from Lesotho and Swaziland). Of this, 13 911x106m3 can be economically harnessed as usable yield
(this includes usable return flow).
Table 1: Available yield in year 2000
Source: DWAF (2002a)
Source Million m3/a
Surface water 10 928
Groundwater 1 042
Usable return flow 1 941
TOTAL 13 911
1
The IPCC defines climate variability as “variations in the mean state and other statistics (such as standard
deviations, the occurrence of extremes etc.) of the climate on all temporal and spatial scales beyond that of
individual weather events. Variability may be due to natural internal processes within the climate system
(internal variability) or to variations in natural or anthropogenic external forcing (external variability)”. In
contrast, climate change “refers to the statistically significant variation in either the mean state of the climate or
in its variability, persisting for an extended period (typically decades or longer). Climate change may be due to
natural internal processes or external forcings, or to persistent anthropogenic changes in the composition of the
atmosphere or in land use.” (IPCC 2001).
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 3
The total amount of water required for 2000 was 13 280x106m3, a figure close to availability limits
(DWAF 2002a). Agriculture is by far the largest user of water, as shown in Figure 1, while urban
and rural domestic requirements make up 25% and 4% respectively.
Figure 1: Water demand for 2000 per sector
Source: (DWAF 2002a)
2.2 Equitable access to water
The issue of access to water is not always determined by scarcity. This is often cited as the reason,
but poor access to water could also be due to political or economic policies. People who do not have
access to water are mostly the marginalised – geographically, economically and socially.
According to the 2001 census, 84% of South Africans have access to piped water, 32% directly into
their homes (SSA 2003). A large percentage of those without access to clean water live in the
historically disadvantaged rural areas, specifically in the previously demarcated ‘homelands’.
The amount paid for water is usually a very small fraction of the household’s disposable income.
However, social or political reasons may require that pricing of water for low consumers should be
subsidized. A commonly observed approach in developing countries is to make use of a rising block
tariff, where the first tier of 6-8 kl/person/month is subsidised or free and the following tiers have
increasingly higher rates (Warford 1997). In South Africa, most municipalities offer this rising block
tariff scheme to all consumers.
To address the issue of affordability, the South African Government committed itself to providing a
life-line tariff, implemented by local authorities, amounting to about 6 000 litres per household per
month free. This after a severe cholera epidemic in 2000 in several provinces and towns, including
parts of Johannesburg. The outbreak was linked by some to the Government’s policy of full cost
recovery for water and the ensuing lack of access to water in sufficient quantity and quality by the
poor (Budds & McGranahan 2003). This policy however does not cater for those individuals who are
not connected to water systems such as people living in rural areas who still have to fetch water from
springs and rivers.
Beyond scarcity and access, water security is also about risk and vulnerability2. One key such risk is
that of projected climate change impacts. Climate change poses a threat to water security for many
of the poorest countries and households. Of course, this threat is not limited to poor countries,
wealthy countries will also experience the impact of changing climate and weather patterns.
However, it is poor people and countries which lack the financial resources to reduce the risk
through firstly preventative action, and secondly through adaptation to impacts or restoration if
damage is inflected by extreme weather events.
2
Vulnerability: “is the degree to which a system is susceptible to, or unable to cope with, adverse effects of climate
change, including climate variability and extremes. Vulnerability is a function of the character, magnitude and rate of
climate change and variation to which a system is exposed, its sensitivity and its adaptive capacity” (Falkenmark et
al. 1989).
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 4
3. Projected climate change scenarios for South
Africa
According to the IPCC (IPCC 2001), global surface temperature is estimated to have increased by
0.6°C since the late nineteenth century, with the 1990s being the warmest decade on the instrumental
record. Most of the 0.6°C increase occurred between 1910 and 1945 and since 1976. Mean daily
surface minimum temperatures appear to be increasing at a faster rate than maximum temperatures
(0.2°/decade versus 0.1°C/decade). Superimposed on these changes are seasonal, annual and inter-
annual variabilities, producing a complex climate variability and change signal.
Climate change studies inherently have to consider the significance of uncertainty. This does not
mean that there is no confidence in the understanding, or that the understanding is not certain enough
to allow for the development of appropriate adaptation strategies and policies for resource
management. Rather, current research would suggest that the political and planning response is
lagging the understanding of climate change. Four sources of uncertainty currently limit the detail of
the regional projects (Midgley et al. 2005):
a. Natural variability: Due to the finite historical records from which the range of natural
variability at different scales of time and space has been defined, it is not possible to set the
definitive limits of natural variability nor to establish how much of the change in variability
is due to anthropogenic factors.
b. Future emissions: Much of the projected change is dependent on how society responds to
reducing emission of greenhouse gases.
c. Uncertainty in the understanding of the climate system: Current understanding of the
regional understanding of the regional dynamics of the climate system of the African sub-
continent is limited.
d. Downscaling: This is the development of regional scale projections of change from the
global models, which introduce an uncertainty that limits the confidence in the magnitude of
the project change, although the pattern of change can be interpreted with greater certainty.
Uncertainty associated with the global and downscaled regional circulation models (GCM and
RCM) estimates of future changes in rainfall is much higher than with changes in temperature, since
precipitation is dependent on a larger number of climatic variables.
The projected impact for southern Africa can best be summarised as follows (Hewitson & Crane
2006):
1. The downscaled rainfall scenarios project a general drying in most seasons in the SW parts
of the Western Cape, particularly during MAM and JJA and in line with a shorter winter
rainfall season. In summer, and autumn the northern and eastern regions of the country are
likely to become wetter, especially over regions of steep topography around the escarpment
and Drakensberg.
2. The projected changes in the intensity and frequency of precipitation events are less certain.
3. Temperature is likely to continue to increase across the country, with the greatest increases
towards the interior, and strongest increase in the daily minimum. This is likely to further
reduce soil moisture content in the dry regions.
4. Vulnerability to climate variation and change in the
water sector
The variability in rainfall is pronounced over southern Africa. This makes planning for times of
drought difficult, yet important, since with large interannual variabilities there are no rainfall
guarantees. The impact of climate change on water resources, including groundwater, acts through a
modification of the water balance, ranging from the micro to the macro-scale. This includes factors
such as surface conditions, the soil column, aquifers and catchments (Braune 1996).
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 5
In a recent review of the current hydroclimatic “landscape” in southern Africa, Schulze motivates
that there is a highly variable and highly sensitive natural hydrological system over southern Africa
because of the following (Schulze 2005):
1. Even when considering average present climatic conditions, we already live in a high risk
hydroclimatic environment in southern Africa
2. An already high inter-annual rainfall variability is amplified by the natural hydrological
system
3. Intra-annual variabilities of hydrological responses are even higher than inter-annual ones
4. Different components of the hydrological system differ markedly in their responses to
rainfall variability
5. Streamflow variability is high in individual external sub-catchments
6. Land use change by intensification or extensification of biomass often increases flow
variability because it changes the partitioning of rainfall into stormflow and baseflow
components
7. Degradation of the landscape can amplify further any hydrological responses, especially
higher order responses.
The South African Country Study on Climate Change found that when running GCM and ACRU3
models, runoff was found to be highly sensitive to changes in precipitation. Groundwater recharge
was found to be even more sensitive (Kiker 2000). Therefore, changes in the climate system will
affect hydrological systems and water resources.
Groundwater in South Africa usually occurs in secondary aquifers and normally soil cover is
shallow. Recharge of the aquifer depends on its type. Some are more responsive to rainfall and
recharge is closely linked to higher and persistent rains; others, such as deep aquifers, are slow to
respond and require consistent rain over a period of time (Visser 2004). Studies by Kirchner (1991)
have shown that before any recharge takes place, a rainfall and soil moisture threshold must be
overcome. The bulk of the recharge takes place in the years in which the average annual
precipitation is exceeded and during periods of high rainfall intensity. It stands to reason, then, that
the areas that are dependent on groundwater will be most vulnerable to decreases in rainfall and/or
variability. In addition, low storage aquifers are the most vulnerable to changes and variability in
recharge. This is the situation in 90% of the country (Braune 1996).
5. Methodological approach to water planning for
climate change impacts
In order to illustrate the effects of climate change impacts on water resource planning and
management in small municipalities, a case study was developed for a small town in the Western
Cape.
After interviewing practitioners in the water sector, it became evident that the issue of available
historical records was key (Murray 2006; Groenewald 2007; Visser 2007; Woodford 2007). Whilst
there are small towns with chronic water problems, the author decided to use the small town of
Bredasdorp (population of 13200 in 2005) as an illustrative example because of the availability of
rainfall records, groundwater extraction records, consumption records, and climate projection
information for the catchment area.
The approach advocated in this paper to plan for the climate impacts on local water resources for
Bredasdorp includes establishing the following four key components:
1. Future water use and demand.
2. Future projected climate change.
3. Future climate impacts on water resources.
4. Future water resource management strategies.
3
ACRU - Agricultural Catchments Research Unit within the Department of Agricultural, Engineering of the
University of Natal in Pietermaritzburg, South Africa
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 6
5.1 Forecast of future water use and demand
Population growth introduces some uncertainty to the projection of future water use and demand.
Researchers differ on their projections of future population growth in South Africa. More recent
estimates are much lower than those done before 2000. Dorrington et al, for example, have included
the consequences of HIV/AIDS in their projections, which results in a growth rate of less than 0.5%
by 2015, whereas the those projections made by Roux in 1998, are in the region of 1% (Roux 1998;
Dorrington et al. 2006).
Table 2: Project annual population growth
Source: Business Futures 1998 and Dorrington et al. 2006
Years Projected annual population growth rates
Business Futures 1998 CAR (2006)
(Roux 1998) (Dorrington et al. 2006)
1999-2015 1.92 % 1.6-0.4 % (0.95 %)
2015-2030 1.06 % 0.4 %
Further, water use patterns differ markedly between urban and rural areas. Rates of water access are
much higher in urban areas and urban population growth rates for earlier periods were substantially
higher than rural areas. Urban population growth from 1946-1970 was 3.45% per year and 3.09% for
1970-1996 (SACN 2004). Overall, this gives a picture of a growing population, but growth slowing
down to lower rates. Further, there have been some suggestions that rural populations have peaked
and will stabilise or even decline due to rural-urban migration (Calitz 1996). This will put further
pressure on urban centres due to migration.
Given this uncertainty associated with population growth, the current trend is to simply multiply the
historic consumption per capita by the historic population to establish the estimated future demand
for water. This is not ideal and also this method does not take into account the increased use of water
as the population becomes more affluent and the level of poverty is reduced. The approach taken in
this paper assumes that no inroads are made into the share of poorer households. Ideally the forecast
should be spatially and sectorally disaggregated and be dependant on a realistic population and water
demand growth figures.
5.2 Projected climate forecasts
Despite steady improvements in the various global and downscaled regional circulation models, the
results from the various models give varying results. No one model is right or wrong, and all project
a plausible time evolution of the climate. The models are in a high degree of consensus over the
direction of change, and how the large-scale circulation features respond to greenhouse gas forcing,
but ascribing probabilities to a magnitude of change is still problematic. Therefore, for planning and
developing response strategies, it is best to consider the bounds of the envelope of possibilities, and
consider the median4 of this envelope. For this study, the median of seven GCMs for 2045-2064 was
scaled linearly to 2035 to establish the percentage change in precipitation as compared with the
model control period of 1980-1999 (Hewitson & Johnston 2006).
The projected percentage change in rainfall was then applied to the mean annual precipitation
(MAP) to establish to potential monthly and annual rainfall in 2035.
5.3 Establish the climate impact on water resources
Both run-off and ground water recharge are affected by a change in rainfall. The extent to which this
occurs has in this paper been called the climate impact factor (CIF). Therefore the expression that
can be used to determine future water resources under a future climate change scenario is as follows:
FS = CS (1 + CIF ) Eq 1
4
Median is used because sometimes a data point is an outlier, and would skew the results if using the arithmetic
mean.
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 7
Where: FS = future supply (kl)
CS = current supply (kl)
CIF = climate impact factor (%)
If the CIF is positive, then we can expect an increase in future water supplies under future climate
impacts due to an increase in rainfall, and visa versa. The CIF can be calculated for either
groundwater or surface run-off. By using this factor, the future available water can be established.
The shortfall or surplus can established by comparing the available future water supplies with the
expected future demand. Where the demand exceeds the projected available supply, additional
sources should be sought and demand-side management (DSM) measures should be introduced. This
is discussed further in the next section.
5.4 Water resource management strategies
Climate impacts are transforming the nature of global as well as local water security, firstly through
climate variability and secondly in the future through projected climate change impacts. Callaway
(2004) argues that there are more conceptual similarities than differences between the adjustments
that are made to cope with climate variability and those made to adapt to climate change. The
obvious similarity is that the aim of both types of actions is to avoid meteorologically induced
damages when predicting them is subject to some error. Both actions have the potential to improve
society, whilst making decisions under some risk, both involve reallocating scarce resources to make
the adaptive adjustments. The major difference, according to Callaway, between variability and
change is that historical records are more reliable for planning for variability than the reliability
attached to climate prediction models. The variability in the existing climate is much easier to plan
for than the variability associated with alternative climates.
As stated previously, global temperatures are rising, with some climatic models suggesting that this
could cause a decrease in runoff in South Africa, spreading progressively from west to east during
the next few decades. A key challenge will be the reconciliation of water demand and supply both
for the medium and long terms. Short-term responses might be seen as coping strategies, whereas
longer-term actions that help to deal with future variability could be collectively called adaptation
strategies. By applying filtering tools which are both qualitative and quantitative (such as multi-
criteria analysis, cost benefit analysis, cost effectiveness analysis) the responses to climate variability
can be evaluated for their long-term suitability to ensuring resilience to future climate impacts, and
thereby ensuring that the local development goals are achieved (as is illustrated in Figure 2).
Climate Variability Climate change
Periodic drought
Short-term water Frequent water
shortages shortages
short-term Filter Long-term
responses resilience
Meeting basic requirements
& development goals
Figure 2: Diagrammatic view of the linkage between climate variability and climate change
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 8
Examples of qualitative filtering criteria are listed below (Mukheibir & Sparks 2006):
• potential for additional yield/savings;
• technology requirements;
• level of capital expenditure;
• impact of additional running costs;
• potential for local employment;
• availability of local capacity to implement;
• acceptability to users;
• impact on the local environment;
• potential for long-term applicability.
Quantitative assessments would include:
• capital and operational costs;
• volume of water saved or provided;
• unit costs R/kl.
It still remains standard practice, however, to project water use on the basis of population and other
factors, but not as a function of climate. Detailed weather records are available for most parts of
South Africa and the supply sources are designed to cope with drought events that have long return
intervals. Large supply facilities therefore have excess capacity at nearly all times, but smaller
systems do not usually have such large margins of safety built in, because of resource and financial
limitations.
Boland (1998) provides some reasons for why conventional planning does not currently consider
climate change:
• In the past, lengthy departures from “normal” weather conditions were viewed as random
deviations around a stationary mean.
• Conventional planning practice leads to large infrequent capacity expansions. Any trend
resulting from climate change would simply cause a change in the timing of the next
capacity expansion, without a loss of reliability.
• Water supply planning based on a static climate assumption, have worked reasonably well
in the past and hence there is no convincing evidence that they will not work in future.
Recently supply options have not been as prevalent and many more utilities are introducing DSM
strategies. Reducing demand can increase excess in supply, thereby creating a greater margin of
safety for future drought periods. It is calculated that the total opportunity in reducing water demand
in the water services sector is approximately 39% of the total existing demand (DWAF 2000).
Examples of current urban water supply management options is listed below (Mukheibir & Sparks
2006). These are, however, not being implemented in a structured and planned manner across all
municipalities.
• Artificial groundwater recharge.
• Conjunctive use of surface and groundwater.
• Desalination.
• Dry sanitation systems.
• Dual flush toilets.
• Education programmes.
• Local water resource management and monitoring.
• Rainfall enhancement.
• Rainwater harvesting.
• Re use of grey water.
• Reduction of leaks programmes.
• Regional water resource planning.
• Saline water for toilets.
• Standby relief under critical conditions eg tankering of water.
• Tariff structures - i.e pricing.
• Water restrictions.
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 9
Drawing on the outputs of the case study by Mukheibir and Sparks, the obstacles and limitations to
implementing these strategies successfully at a local municipal level were identified. Two key ones
stand out:
1. Local capacity: The most notable issue affecting the viability of these strategies is the perceived
lack of local capacity to implement the strategies. The former Director-General of DWAF, Mike
Muller, has stated that there is a severe shortage of qualified water managers in small-to-
medium-sized municipalities which has resulted in 63% of municipalities not complying with
the drinking-water quality standards. There is an urgent need for formal training in this sector
(Venter 2005).
2. Financial: This is further exacerbated by the low financial resource base to cover the capital and
running costs of most of the strategies. Local government competes for nationally allocated
funds for capital expenditure. Running costs are mostly covered from local revenues, which for
the smaller and remote local municipalities, are insufficient to ensure water security at this level.
In addition political buy-in for some of the strategies such as water restrictions and dry sanitation
will need to be obtained through education programmes, however these also require human and
financial resources.
6. Water balance assessment for Bredasdorp
6.1 Present water resources
In 2002 Bredasdorp obtained its bulk water supply from the following surface and underground
water resources:
Groundwater:
Five production boreholes (BD1-5) were developed 1985 with a safe yield of 480 000 kl/year. The
boreholes were located in the quartzitic sandstone of the Table Mountain Group to the south of the
town. The maximum to be extracted over the summer peak period is 59 000kl/month (Toens et al.
1998). During the drought conditions in 2001 BD6 was developed and in 2002 testing pumping of
BD7, BD8 & BD9 was done. The total annual available equipped yield for the groundwater in 2001
was 792 000 kl.
Table 3: Actual abstraction
Source: Afri-Coast (2003)
Year BD1-BD6 BD7-BD9 Average rate Available yield % of available
(kl) (kl) (kl/d) (kl) yield
1999 418 185 0 1 146 480 000 87.1 %
2000 216 965 0 595 480 000 45.2 %
2001 444 037 131 148 1 576 792 000 72.6 %
Surface water:
The Klein Sanddrift Dam, situated five kilometres to the west of the town, has a capacity of
455 000kl and was commissioned in 1997 to capture the “surplus” run-off from the catchment. The
98% yield of this source is 440 000 kl/year (1205 kl/day) (Ninham Shand 1998). In 2001 the water
level dropped to below 33% of full capacity (Afri-Coast 2003).
Table 4: Water abstracted from Klein Sanddrift Dam
Source: Afri-Coast (2003)
Year Bredasdorp Farmers Average total
1999 425 314 118 017 1 488
2000 388 447 159 581 1 501
2001 221 746 131 083 967
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 10
The Uitvlugt springs supplies water via a pipeline over 11km to the Vleikloof dam. The estimated
sustainable yield is 160 000 kl/annum (438 kl/day). The Vleikloof dam has a capacity of 295 000 kl
and is useful in meeting peak demands (Ninham Shand 1998).
Summary of water supply resources:
As can be seen in Table 5, the available potential equipment water annual supply is 1 257 Ml, of
which 63% is dependent on groundwater.
Table 5: Summary of Bredasdorp water supply (in kl/year)
Source: Afri-Coast (2003)
Year Dam Spring Groundwater Total
Potential 305000 160000 792000 1257000
1999 425314 148305 418185 991804
2000 388447 236354 216965 841766
2001 221746 233716 575185 1030647
6.2 Future supply options
No surface water is available for further exploitation, but three further underground aquifers were
identified as potential supply options. This will increase the dependence of the town on groundwater.
The De Duine West option is much more expensive in terms of unit cost since it is located 22 km
from Bredasdorp.
Table 6: Future groundwater supply options
Source: Toens et al. (1998):
Underground Supply option Estimated safe yield Real cost in 2005 Unit cost
(R/kl)
Golf course compartment 260 000 kl/year 4 870 000 18.75
Sanddrift/Napier compartment 800 000 kl/year 12 500 000 15.65
De Duine West 720 000 kl/year 33 725 000 46.85
Total 1 780 000 kl/year 51 095 000 28.71
6.3 Water demand
Water demand growth rate until 2008 was estimated to be 2.3% based on the historical growth (Afri-
Coast 2003). Based on this assumption, the annual demand in 2035 will be 1 963 ML.
If we consider the demand for water against the existing supply (see Figure 4), we see that a shortfall
can be expected around 2016 (see Figure 5).
6.4 Unaccounted for water – Losses
In 2002 water losses were in the order of 15% in Bredasdorp. In 1999 this was as high as 24.2%.
This was reduced by implementing an improved water management programme (Afri-Coast 2003).
Table 7: In-line water losses
Source: Afri-Coast (2003)
Year 1999 2000 2001 2002
% losses 24.20% 20.40% < 15% ~ 15%
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 11
7. Impact of climate change on water resource
management
7.1 Projected annual precipitation for the Bredasdorp catchment
Most methods used to estimate future water supply resources assume that climate is stable and that
mean precipitation does not change over time (Kirchner 2003). However, as can be seen by Figure 3,
the mean annual precipitation (MAP) for Bredasdorp is recorded as 500mm (Ninham Shand 1998),
but the average for the 25 year period (1980-2004) is 481 mm per annum (3.8% less), which would
indicate some drying already.
In the Western Cape region, precipitation is forced by large scale circulation but is significantly
modulated by the topography. Future projections for winter are for continued winter drying in the
west in both lowland and mountain regions. The winter season is likely to shorten, while summer
will be characterised with drying in the west, with slight wetting in the north and east (Hewitson &
Johnston 2006).
For the Bredasdorp catchment, the projected precipitation for 2035 decreases by 8% as compared
with the mean annual precipitation (MAP; Hewitson & Johnston 2006; see Figure 3). Hydrological
and engineering planning usually get done using the MAP. In future therefore, projects for
Bredasdorp with a lifespan of 30 years should be planned for using 8% less precipitation than the
MAP. The effect of climate change on rainfall amounts and patterns of rainfall, particularly on
intensity and increased evaporative losses due to higher temperatures, may be more important in its
impacts on run-off and groundwater recharge than the rising temperature alone.
750
Annual Precipitation
700 25 year average('80-'04)
historical MAP
650
projected for 2035
600
550
mm 500
450
400
350
300
250
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
Figure 3: Annual precipitation for Bredasdorp
7.2 Run-off estimation
When considering the impact of reduced precipitation on run-off, work done by de Wit and
Stankiewicz is of interest since it shows that for a hypothetical 10% reduction in rainfall in sub-
Saharan Africa, regions with 1000 mm MAP would experience a 17% reduction in catchment run-
off and for areas that have 500-600mm MAP would have 50-30% less run-off respectively (de Wit
& Stankiewics 2006). While using an average precipitation reduction for the continent is a gross
simplification and projections using climate models on a regional basis should be used, the related
run-off reductions are of interest here.
Studies by Chapman (2007) on historical data reveal that the percentage reduction in run-off can be
up to 2.5 times greater than the reduction in precipitation. He cautions that the use of simple trend-
lines to arrive at these figures can be misleading, and that the absolute figures should not be used.
However the clear trend that a percentage reduction in precipitation results in a greater percentage
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 12
reduction in runoff is important. A modelling study conducted by Bari et al of the impact of climate
change on run-off in Western Australia revealed that for an 11% decrease in precipitation by 2050, a
31% reduction in annual catchment yield would likely be experienced (Bari et al. 2005). What these
studies reveal is that the magnitude of the run-off reduction is two to four times greater than the
reduction in the precipitation. However, historical data is required to establish the direct relationship
for a site specific analysis.
Unfortunately, historical run-off data is not available for the Bredasdorp catchment. However, based
on the above, we can assume that for a projected 8% reduction on a MAP of 500mm by 2035 a
reduction in run-off would be in order of 50-30%. Therefore, for the purposes of this study we have
chosen a conservative reduction of 30%, i.e the CIF would be -0.30.
Therefore, FSrunoff = CSrunoff * (1 – 0.3)
7.3 Recharge estimation
Groundwater recharge estimation is a key factor in determining the suitable management of
groundwater resources. A wide range of techniques is used for estimating groundwater recharge as
well as direct and indirect measuring methods with which recharge is calculated, inferred or
simulated (Beekman & Xu 2003). Using long-term records of past climate-recharge interactions is
difficult to apply in southern Africa, since there are very few locations where records of climate and
groundwater have been kept in sufficient detail to allow for this analysis (Selaolo et al. 2003).
Various studies of Bredasdorp and the Table Mountain Group (TMG) aquifers provide varying
information on the recharge of the aquifer. Toens et al (1998) obtained an average recharge of 19.6%
from various reports citing other locations. This is supported by recharge calculation of 17.4% at
neighbouring Struisbaai using cumulative rainfall collectors (Weaver & Talma 2005). Duah et al
(2006) have estimated that recharge for TMG aquifer ranges from 0.28% to 12.6 % with an average
of 30mm/year. For the purposes of this study, a recharge of 19.6% for MAP of 500mm is used as
calculated by Toens et al.
Cave et al argue that short-term predictions of climate change are not necessarily more accurate than
long-term ones. Human-induced climate change will occur against a backdrop of natural climate
variability. Climate forcings in the short-term are weak in comparison to long-term ones and so the
short-term signal is more difficult to separate from that of climate variability (Cave et al. 2003).
In general, decreasing precipitation will cause decreasing recharge and hence depletion of the
groundwater resources. Changes in recharge will result from changes in rainfall as well as a change
in the timing of the rainfall season. Rainfall-recharge relationships may therefore be used in a first-
cut assessment of climate change impacts on the groundwater resources.
A comparison of studies conducted in many disparate locations using different methods revealed that
in areas with annual precipitation less than 500mm/year, large differences exist between the recharge
values found. Recharge becomes negligible for precipitation lower than about 400mm/year
(Beekman et al. 1996). Beekman et al. developed the following equation:
y = 148 ln (x) – 880 Eq. 2
Using this equation, changes in longer-term groundwater recharge due to predicted long-term
changes in annual precipitation can be calculated. Applying this equation to a MAP of 500mm
would result in a recharge of 39.8mm/year. If reduced by 8% to 460mm, recharge would be
27.4mm/year. This would equate to a 31% reduction in groundwater recharge. Therefore recharge is
reduced by a greater percentage than the reduction in rainfall and the groundwater storage would be
thus accordingly affected.
The recharge percentage MAP would change from 19.6% to 13.5%, a reduction of 31%, i.e the CIF
would be -0.31.
Therefore, FSrecharge = CSrecharge * (1 – 0.31)
This would mean that groundwater abstraction would also need to be reduced by 31% so as not to
dewater the aquifer.
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 13
7.4 Impact on water supply
To maintain the current levels of supply till 2035, an additional 430Ml/year (marked A in Figure 4)
will be needed under projected climate change conditions, as compared with the supply under
normal climate conditions. This equates to 31% of the current supply under normal conditions (i.e.
A/Total Supply). In order to meet the projected demand for water in 2035, this climate-induced
water supply decrease will need to accommodated. This climate-induced shortfall is also 54% of the
unmet demand of 2035 under normal conditions (i.e. A/B). This means that by 2035, planners will
need to plan for 54% more water than they would have under normal climate conditions.
2,000,000
Total Supply:
1,800,000
Projected consumption
1,600,000 B
Projected supply under CC
1,400,000
Kl 1,200,000
1,000,000 A
800,000
600,000
400,000
86
90
94
98
02
06
10
14
18
22
26
30
34
19
19
19
19
20
20
20
20
20
20
20
20
20
Figure 4: Water supply and demand graph under historical and climate change
conditions for Bredasdorp
Further, under normal demand projections, the current water supply system experiences a shortfall
by 2016. The projected climate change impacts (i.e. 8% reduction in MAP) will induce a shortfall by
2009, seven years earlier. This will have infrastructural and financial implications and is discussed
later.
300,000
under normal climate
200,000
under climate change
100,000
Kl 0
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
-100,000
-200,000
-300,000
-400,000
-500,000
Figure 5: Timing of projected water supply shortfall for Bredasdorp
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 14
8. Analysis of potential water resource management
strategies under climate change impacts
A number of supply- and demand-side options are investigated to illustrate the impact of climate
change on water resource management for the Bredasdorp case. These strategies have not been
proposed or adopted by the planners of Bredasdorp.
8.1 Selected supply-side options
To meet the estimated demand by 2035, an additional 380Ml/year will be needed under projected
climate change conditions, as compared with the supply under normal climate conditions. This is
54% more than under normal climate conditions. As can be seen in Figures 6 and 7, three additional
sources of water supply will be needed under climate change conditions to ensure the demand is met
in 2035.
2,000,000 Total Supply:
Projected consumption
1,800,000
Future Supply:
1,600,000
1,400,000
Kl 1,200,000
1,000,000
800,000
600,000
400,000
86
90
94
98
02
06
10
14
18
22
26
30
34
19
19
19
19
20
20
20
20
20
20
20
20
20
Figure 6: Projected water supply till 2035 under normal climate conditions
2,000,000 Future Supply under CC:
Projected consumption
1,800,000
Projected supply under CC
1,600,000
1,400,000
Kl 1,200,000
1,000,000
800,000
600,000
400,000
86
90
94
98
02
06
10
14
18
22
26
30
34
19
19
19
19
20
20
20
20
20
20
20
20
20
Figure 7: Projected water supply till 2035 under climate change conditions
There is a progressive reduction in available water supplies over the period ending in 2035, which
results in a 31% and 30% reduction in recharge and run-off respectively. In other words, a potential
resource in year 2000 of 400Ml would be reduced to 328Ml by 2016 and further reduced to 306Ml
by 2024, due to the increasing impact of climate change (see Table 8).
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 15
Table 8: Future groundwater supply options commissioned (in kl/year)
Supply Normal climate Climate change conditions
option (present estimated supply) (present estimated supply has been reduced by the annual CIF )
2016 2025 2035 2010 2016 2024 2030 2034 2035
Golf course 260 000 224 178
compartment (260000)
Sanddrift/ 450 000 328355 306311
Napier (400 000) (400 000)
compartment
De Duine 260 800 111 502
West (360 000) (160 000)
Cumulative 260 000 710 000 710 000 220 480 541787 811724 1028711 1 101 084 1 090 200
8.2 Selected demand-side strategies
A sample of possible DSM savings and related unit costs are provided in Table 9. To illustrate the
effect of DSM measures on future water resource management and planning, outdoor water use
restrictions and active leak detection programmes have been analyzed in this example.
Table 9: Examples of demand-side options
(after White & Fane 2002)
Measures Saving Cost
(l/p/d) (2005 R/kl)
Shower head performance standard 8.6 0.01
Active leakage control 7.2 1.71
Clothes washer performance standards 3.5 0.23
Residential indoor audit and retrofitting - with discount 3.4 1.08
Industrial and commercial audits 2.9 2.40
Industrial reuse project 1 2.3 3.02
Price increase (R0.57/kl over 2 years) 1.9 0.01
Outdoor water use restriction 1.8 0.36
Industrial reuse project 2 1.8 3.71
Residential indoor audit and retrofitting - free for low-income 1.5 1.43
Hotel audits 1.3 2.40
Shower head rebate (R56.89) 0.7 0.80
Washing machine rebate (R853.28) 0.4 3.99
Outdoor irrigation system audits 0.3 3.82
Outdoor water use promotion 0.2 2.79
The introduction of the DSM measures results in a delay of 1-2 years for the implementation of the
capital intensive supply options. Based on a projected population of 26150 by 2035, the volume of
water saved for the year in 2035 is approximately 86Ml. The financial implications are discussed
further in the next section.
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 16
2,000,000 Projected demand under DSM
Future Supply under CC:
1,800,000
Projected demand
1,600,000 Future supply under CC & DSM
1,400,000
Kl 1,200,000
1,000,000
800,000
600,000
400,000
86
90
94
98
02
06
10
14
18
22
26
30
34
19
19
19
19
20
20
20
20
20
20
20
20
20
Figure 8: Projected water supply till 2035 under climate change and DSM conditions
9. Financial assessment of strategy options
Financial cost indicators are ideal for qualitatively assessing the various supply and demand
reducing options, since for local municipalities financial capital is one of the key resources at their
disposal.
In order to assess the financial implications of meeting the projected demand for 2035 under climate
change conditions, the real present day investment cost (assumed here to be as at 2005) of additional
supply has been compared with that under normal climate conditions. The capital costs to secure the
groundwater resources was obtained from relevant engineering reports (Toens et al. 1998). The
influence of the change in revenue and running costs have been assumed to be negligible, since the
projected demand is the same under both scenarios. Treatment costs will therefore remain constant,
whilst pumping costs may be slightly higher due to low water tables and longer pumping times.
This analysis reveals that approximately an additional R30 000 0005 will be required. This is
approximately 3.5 times the capital expenditure under normal climate conditions. In other words, for
this case study, a projected decrease of 8% MAP by 2035 will result in a 251% increase in cost.
Table 10: Capital cost comparison of ensuring water supplies under climate change conditions
No CC Under CC change
Impact by 2035
Real 2005 costs of additional supply 11 902 128 41 728 505 251% 3.51times
Additional annual water supplied by 2035 710 000 1 090 200 54% 1.54 times
Real costs/additional water required by 2035 (R/kl) 16.76 38.28 128% 2.28 times
Impact by 2020
Real 2005 costs of additional supply 4 869 999 11 120 780 128% 2.28 times
Additional annual water supplied in 2020 260 000 501 600 93% 1.93 times
Real costs/additional water required by 2020 (R/kl) 18.73 22.17 18% 1.18 times
Cost-effectiveness analysis is a specialised version of traditional cost-benefit analysis. All costs of a
portfolio of strategies are assessed in relation to a policy goal that represents the benefits of the
strategies, and all other impacts measured as positive or negative costs. The policy goal in this case
is ensuring municipal water supply. The unit cost comparison has been calculated by dividing the
5
Approximately US$ 4.25 million (using R:$ = 7:1)
ENERGY RESEARCH CENTREImpact of climate change on municipal water resources 17
real cost by the actual volume of additional annual water required in 2035, expressed in Rands per
kilolitre. Therefore, under the normal climate scenario, the cost of meeting the 2035 demand is
R16.5/kl as compared with R38/kl under climate change conditions. This is 2.5 times higher, as is
graphically illustrated in Figure 9. This is less than the real cost ratio, since the additional volume of
water required has increased by 54% (i.e. 1.54 * 2.28 = 3.51).
If we consider the climate impact by 2020, we note that the cost comparison for the year 2020
equates to an increase of 128% (2¼ times). This means that the cost increase due to climate change
doubles from 2020 to 2035. This is due to the fact that over time the impacts of climate change on
run-off and recharge become more pronounced, and hence water becomes more difficult and
expensive to secure to meet the projected demand. This is further illustrated when considering the
unit costs. The unit cost for water in 2020 was R22/kl – just more half that for 2035.
2.5
normal climate
2.0 climate change
1.5
Times
1.0
0.5
0.0
2020 2035
Figure 9: Magnitude of the increase in unit cost due to climate change over time
The impact of implementing DSM strategies is that the saved water results in a delay in the
implementation of supply infrastructure, and hence saves capital investment costs. For example,
there is a 7.6% increase in available water due to the two DSM options by 2035 i.e. 86 Ml. This
results in a delay of the implementation of supply options by 1-2 years during the period of study,
causing a net decrease in the investment cost of 7.8%, and a reduction of the unit cost by 8.1%. It is
acknowledged that the reduction in demand may result in reduced revenues, but this will in all
likelihood be offset by the reduced operational costs.
Table 11: Capital cost comparison of ensuring water supplies under climate change conditions
and DSM
Impact by 2035: Under CC Under CC & DSM Change
Real 2005 costs of additional supply 41 728 505 38 477 070 -7.8% 0.92 times
Additional annual water supplied by 2035 1 090 200 1 007 400 -7.6% 0.92 times
Additional annual water saved through DSM by 2035 0 86 094 7.6%
Total additional water secured by 2035 1 090 200 1 093 306 0% 1.00 times
Real costs/additional water required by 2035 (R/kl) 38.28 35.19 -8.1% 0.92 times
ENERGY RESEARCH CENTREYou can also read
